C2PA AI/ML Assertion Decoder

Decodes AI/ML provenance assertions off a C2PA manifest's assertion array: ALL entries in every c2pa.actions/c2pa.actions.v2 assertion (a manifest can carry a created -> edited -> published history; this reads the whole chain, not just the first action), the full IPTC digitalsourcetype NewsCodes vocabulary (raw code always surfaced, an unrecognized code is never silently dropped), and c2pa.ai_training / c2pa.ai_generative_training training-and-data-mining opt-out assertions (training_mining_opt_out: true/false/not_asserted, read only from an explicit boolean, never inferred). Reports assertions only -- never adjudicates whether an assertion is true, and absence of an AI/ML assertion means nothing. Reuses art-123's assertion-array input shape as a sibling stage; zero edits to art-123 or art-361. Zero network calls, zero PII.

Tool URL: https://ainumbers.co/chaingraph/art-597-c2pa-aiml-assertion-decoder.html